Open Access
Add to BibSonomyAbstract
We report on an experimental study of the effect of atmospheric turbulence on laser induced breakdown spectroscopy (LIBS) measurements. The characteristics of the atmosphere dictate specific performance constraints to this technology. Unlike classical laboratory LIBS systems where the distance to the sample is well known and characterized, LIBS systems working at several tens of meters to the target have specific atmospheric propagation conditions that cause the quality of the LIBS signals to be affected to a significant extent. Using a new LIBS based sensor system fitted with a nanosecond laser emitting at 1064 nm, propagation effects at distances of up to 120 m were investigated. The effects observed include wander and scintillation in the outgoing laser beam and in the return atomic emission signal. Plasmas were formed on aluminium targets. Average signal levels and signal fluctuations are measured so the effect of atmospheric turbulence on LIBS measurements is quantified.
©2009 Optical Society of America
1. Introduction
Laser-induced breakdown spectroscopy (LIBS) represents one of the most promising approaches for addressing the emerging needs of industry, environmentalists and security specialists by providing a rapid and field deployable technology for identification and quantitative analysis of distant objects. TELELIBS systems analyze targets at several tens meter distances by transmitting laser beams directly through the atmosphere. This technology offers many features, principal among them being the capability for the analysis of the elemental composition of distant objects. This is a unique capability of LIBS among the remote analytical technologies. TELELIBS is compatible with a wide range of applications, and is sufficiently flexible to be easily implemented in a variety of working scenarios such as production plants [1–3] and urban environments [4].
TELELIBS systems have been described using a variety of optical architectures. Several of the first attempts used single optical elements for laser focusing. With these simple designs the maximum distance reached was about 5 m [5]. The TELELIBS analysis of solids at large distances using telescopes was first reported in the LIBS 2000 Conference in Pisa [6] and subsequently published elsewhere [7]. A couple of refracting telescopes were used for laser beam delivery and plasma light gathering. The maximum tested range was 45 m. Soon after this report increased distances were reached using reflecting telescopes for the return path [8]. Large mirrors are advantageous for this application as they are less costly than lenses of the same size and are free from chromatic aberration, while UV-enhanced aluminium mirrors provide excellent reflectivity from the UV to the NIR. Longer distances were reached by Grönlund et al. in 2006 [9]. LIBS images were obtained by a mobile lidar system at a stand off distance of 60 m using a large computer-controlled mirror. The same year, a field-deployable LIBS-based system for stand-off measurements was presented [10]. In this paper main factors influencing LIBS performance at stand-off distances were outlined. However, no account for atmospheric effects was presented in these reports.
TELELIBS is strictly a line-of-sight technology, i.e., the laser transmitter and the sample are able to “see” each other. In order precisely to identify and distinguish an analyte placed at a large distance from the sensor, one must accurately point the laser beam over the target, and recover the plasma emission with the highest possible efficiency. With increased distances, however, transmission of both the laser beam and the returning plasma emission through the atmosphere is subject to considerable angular fluctuations due to natural turbulences in air. In the absence of attenuating elements such as large water particles of rain, the atmosphere is best modeled as a random phase medium that changes with time. To a first order, the atmosphere introduces a random deflection of a laser beam transmitted through it. For example, on a sunny day the rising hot air makes the index of refraction of air goes up with height. Such an index of refraction change near the transmitter tends to deflect the beam causing “beam wander”. The same effect near the receiver causes the beam to appear to have come from a different place. This effect is usually termed as angle-of-arrival fluctuations. The magnitudes of these effects are largely dependent on the index of refraction fluctuations and the propagation distance and affect to a larger extent beams propagating parallel and close to the ground as is the case in many TELELIBS experiments.
The main purpose of this manuscript is to analyze the atmosphere characteristics that affect the signal performance of this LIBS technology when working in an open range. A new TELELIBS system is described which incorporates large scale reflecting optics for laser beam delivery and plasma light collection and a number of subsystems for added functionality such as beam pointing, range finding, and video recording capabilities. A maximum distance of 120 m is examined. Atmospheric turbulence has been studied extensively and various theoretical models have been proposed to describe turbulence-induced degradation of free-space optical communication links [11,12]. This paper addresses the specific effects of atmospheric turbulence on TELELIBS signals and discusses the main variables affecting LIBS detection capabilities.
2. Experimental
Figure 1 shows a picture of the stand off LIBS instrument ready for operation. A high power double pulse Nd:YAG laser system (Quantel Brilliant Twins, 10 Hz, 750 mJ per pulse each, 5.5 ns pulse width) operating at 1064 nm is used as the ablation source. The system can be used in sequential dual-pulse mode by synchronizing the lasers with two pulsed generators (Berkeley Nucleonics) that allow control of the energy and time delay between pulses. However, for the present investigation the two lasers were fully overlapped in time and space to result in a single beam of sum energy per pulse of about 1.5 J at the exit of the beam combining module. In order to help guiding the infrared laser to the target, a diode laser at 635 nm is sent collinear with the excitation beam. The laser was directed towards a telescope by consecutive reflections on five laser mirrors of an articulated arm (Applied Photonics Ltd.) designed to follow the displacements of the telescope when moving in altitude and azimuth. The articulated arm keeps the alignment of the beam from the laser head to the telescope entrance port. To achieve the necessary beam divergence before reaching the secondary mirror of the telescope, a beam expander was used. The beam expander is formed by a pair of antireflective-coated fused silica lenses, whose effective focal lengths are selected considering the teledetection range of the instrument. A dichroic mirror placed at an angle of 45° to the beam expander reflects the laser beam towards the telescope. With the purpose of minimizing time of design and manufacture as far as possible, of-the-shelf components have been chosen. In this way, a 400-mm aperture classical Cassegrain (Optical Guidance Systems) open-truss telescope was used to focus the beam onto the sample. Among the advantages of the Cassegrain design are that it offers an external and very accessible focus and that is free from spherical aberration. The telescope has been fitted with UV-coated optics to enable work in LIBS applications. The expanded laser beam enters the telescope through the back aperture, is directed to the secondary mirror and then reflected to the primary mirror and the sample. In order to reduce the obscuration produced by the secondary mirror, the laser beam is sent slightly off axis while still maintaining the focusing capabilities of the telescope. The available pulse energy after exiting the telescope was 1.2 J. The telescope focuses the laser beam on the target. The beam focusing distance was adjusted by changing the distance between the primary and secondary mirrors. The minimum focusing distance of this telescope is restricted by the maximum travel of the secondary mirror. In this case, the distance is about 20 m. The telescope had a fork-type Dobsonian mount. With this system, the maximum irradiance is attained when the beam is tightly focused on the target. However, the irradiance is much reduced when the beam is expanded. In the proximities of the expanding 400-mm Cassegrain telescope used here, the irradiance can be stimated in 0.2 MW/cm2, assuming that the telescope primary mirror is completely filled with the beam. This irradiance is well below the threshold level required for inducing breakdown in most solids. The irradiance will increase with the distance downfield to reach GW/cm2 levels close to the sample.
Plasma light collected by the telescope is directed through the aperture of the primary mirror to the focal point located behind the dichroic mirror. The plasma image was brought to a spectrometer (Andor SR-303iA) using a 2-m long, 600-µm diameter fiber optic. The spectrometer was fitted with classically ruled gratings of 150, 300, 1200 grooves·mm−1 with blaze at 500, 500 and 300 nm, respectively. Detection of the dispersed light is made with an intensified charge coupled device detector with 1024 x 1024 pixels (pixel size 13.5 µm2, Andor iStar DH740-25F-03).
With the aim of inspecting the shape and size of the plasma formed in the remote target, plasma light can be directed towards a CCD camera (Sony XCL-X700) using two flat 2.54-cm aluminium mirrors instead of being directed towards the optical fiber. One of the mirrors is placed in a moving mount sited at 45 degrees of the visible light; the other one is placed in a fixed mount also at 45 degrees. As additional components, the instrument incorporates a digital camera for on-line observation of the target and its surroundings and a range-finder to accurately measure the distance between the equipment and the sample. All the system is mounted on a wheeled platform fitted with six vibration-dumping mounts. Figure 2 shows an optical layout of the TELELIBS instrument.
Fig. 2 Optical layout of the TELELIBS instrument. (1). Diverging lens, (2). Converging lens, (3). Dichroic mirror, (4). Primary mirror, (5). Secondary mirror, (6). Flip mirror, (7). Folding mirror, (8). Optical fiber, (9). CCD.
A firing range consisting of a 50 m partially closed corridor was available for indoor measurements. Outdoor experiments were conducted during day time on May 2008 outside the firing range. Atmospheric conditions were windy with maximum speed of 30 Km h−1 and frequent wind gusts. The visibility was 10 Km and the temperature outdoor was 18 °C. The horizontal TELELIBS beam was at a height of 1.5 m above the ground and angled approximately 60° to the prevailing wind direction. In order to protect pedestrians from exposure to laser radiation, the area was evacuated appropriately as the energy used in this system is far beyond the threshold irradiance for ocular damage, in particular in the surroundings of the sample. Aluminium plates were used as samples throughout the experiments.
3. Results and Discussion
An optical wave front propagating through the atmosphere will experience a number of alterations due to small fluctuations in the refractive index of the atmosphere (optical turbulence). These random fluctuations are directly associated with microscopic temperature fluctuations caused by turbulent motion of the air due to wind and convection. The fluctuations evolve and move across the beam over time, creating turbulent cells. Although the fluctuations are in general very small (only a few parts per million), a propagating laser passes through a large number of turbulent cells, so their accumulative effect on the transmission is quite large. Atmospheric turbulence is constantly present under all atmospheric conditions. However, while its effects on standoff LIBS measurements is negligible when the range considered is of a few meters, measurements involving ranges of several tens meters can be significantly affected by atmospheric turbulence. For the case of TELELIBS, which implies horizontal path propagation and double passage of optical waves through the atmospheric inhomogeneities (outgoing laser beam and return atomic emission light), stronger effects are to be expected. At least four consequences of atmospheric turbulence are of relevance for TELELIBS measurements. These include (1) a spreading of the laser beam, (2) a random variation of the position of the laser beam centroid on the target (beam wander), (3) fluctuations in the position of the plasma image at the telescope focal plane and (4) irradiance fluctuations affecting both the laser beam transmission and the atomic emission reception. In the following paragraphs these effects are discussed.
Atmospheric turbulence degrades the spatial coherence of the laser beam as it propagates through the atmosphere. This loss of spatial coherence causes the beam to spread and limits the extent to which the laser beam may be focused. As a result, a significant irradiance reduction at some distant object occurs when compared with propagation at short distance. Figure 3 shows the laser beam cross-section obtained with a beam analyzer in the near field just before entering the telescope system (Fig. 3(a).) and the cross-section when tightly focused with the telescope after propagating 50 m along a horizontal path at 1.5 m from the ground (Fig. 3(b).). The effect of atmospheric turbulence is clearly noticed, and may include diffraction effects induced by the focusing optics. The beam in the near field exhibits a typical multimode output, while the focused beam at 50 m shows the difficulties in achieving a tight focus. The irradiance in this case is computed from a circle containing approximately 90% of the observed beam cross section.
Fig. 3 (a) Colimated near field intensity beam cross section. (b) Focused intensity beam cross section after propagating 50 m at a height of 1.5 m above the ground.
3.1 Beam wander effects on LIBS measurements
Beam wander, caused by random deflections of the beam as it propagates, causes the instantaneous center of the laser beam to displace at the target. Statistically, beam wander can be characterized by the standard deviation of the beam centroid displacement along an axis (sW, m). For a focused beam it is given by Eq. (1) [11].
where r is the propagation distance (m) and W0 is the laser beam radius (m). Cn 2 is the atmospheric structure parameter (m-2/3), which is a measure of the strength of the fluctuation in the refractive index of the atmosphere and thus directly related to atmospheric turbulence. From Eq. (1) it is concluded that beam wander should be particularly noticeable for a pulsed beam on long propagation paths in the presence of turbulence. The time scale of these fluctuations is about the time it takes a volume of air the size of the beam to move across the path and therefore beam wander is related to wind speed. Usually turbulence fluctuations are much longer in time than laser pulse length in the nanosecond regime, thus no specific effects associated to the pulse width are observed. Figure 4 shows photographs of Al targets on which an increased number of shots have been delivered when the target was placed at 120 m from the instrument. The maximum wind speed during the measurements was 30 Km·h−1. The wind direction and speed varied during the measurements, while occasional wind gusts were observed. The spreading of the imprints left on the target caused by beam wander is clearly shown. It should be noted that the spreading increases with the number of laser shots. This is certainly to be expected since for increased acquisition times the probability of wind gusts also increase. Considering only the displacement along the horizontal axis in the pictures, the size of the footprint is 0.47 mm for a single shot, 0.81 mm for 20 shots, 1.38 mm for 100 shots and 1.80 mm for 1000 shots. Beam wander also grows with the propagation distance as noted above. For 50 laser shots, the footprint size increases from 2.18 mm at 30 m, to 2.93 mm at 50 m, to 3.74 mm at 120 m. It should be noted that the maximum spread depends on the particular wind speed and direction of the observations, so no general trend can be established for the maximum beam spreading.Fig. 4 Inprints left on an aluminium target at 120 m after (a) 1 shot, (b) 20 shots, (c) 100 shots and (d) 1000 shots.
For comparison purposes, beam deflections were measured inside a firing range where the effect of wind should be less effective. For this purpose a beam analyzer was placed at the target position at two different distances from the instrument. Figure 5 shows the displacement of the beam centroid along the records acquired. The fitted curves are the zero-mean Gaussian distributions. As shown, the scattering of the beam hits is significantly larger at 50 m, with beam displacement standard deviation values of 345.0 µm at 50 m, and 242.3 µm at 30 m. The largest deflection of the beam at 50 m approaches 900 µm in the right wing of the distribution around the centroid average position. The distribution is decidedly asymmetric as responding to the dominant wind direction. It is also observed that larger deflections are more frequent at 50 m. When compared with the data acquired outdoor, the beam fluctuations in the firing range are significantly smaller due to the absence of wind, while fluctuation due to changes in temperature with height and the inherent beam pointing instability of the lasers should account for the deviations observed.
Fig. 5 Displacement of the laser beam centroid along the horizontal axis. The fitting curves mean the zero-mean Gaussian distribution of the deviations as measured at distances of 30 m and 50 m.
A plasma moving on the target as an effect of beam wander will cause changes in the image position of the plasma return signal by changing the angle of arrival on the receiving optics. This effect may significantly change the magnitude and the noise level of the observed LIBS signal if the aperture used to collect the light is small compared to the image displacement. The effect would be observed when either a fiber optics or a spectrometer slit is placed at the telescope focal plane. However, the imaging capabilities of telescopes tend to alleviate the atmospheric effects as illustrated in Fig. 6 . In this figure, registers for the plasma image position in the focal plane were acquired with a beam analyzer and the frequency of deviations is plotted for targets located at 30 m and 50 m. These data were subsequent to the data in Fig. 5. As shown, most images deviate from the central average position less that 20 µm. Also observed is that the deviations from the average position are virtually the same at the two distances tested, with standard deviation values of 7 µm and 8 µm at 30 m and 50 m, respectively. Recalling the results of Fig. 5, laser beam deviations at the target are significantly larger than those observed at the receiver (Fig. 6, note the difference in the scale of the x axis compared with Fig. 5). In other words, laser beam displacements at the target causing the plasma to fluctuate up to 900 µm translates into the detector as only 20 µm image displacement. Although smaller in magnitude, these fluctuations have important consequences on the design of the capturing optics of a TELELIBS analyzer. In the present TELELIBS system, an increase in the diameter of the optical fiber used for image transfer to the spectrometer beyond the image fluctuation width will reduce the effects. Since the size of the image at the detector measured with the beam analyzer is 0.374 mm (1/e2 diameter), a 600 µm optical fiber, as used in our system, is large enough for capturing the full image of the moving plasma and reduce the signal fluctuation. A 50 µm fiber, for instance, would cause excess noise in the observed signal due to image displacement at the telescope focal plane, causing major detrimental effects on the measurement precision.
Fig. 6 Displacement of the plasma image position along the x axis in the telescope focal plane acquired with a beam analyzer. The fitted curves are the zero-mean Gaussian distributions of the deviations.
3.2. Turbulence-induced irradiance fluctuations and measurement precision
Propagation of the laser beam through turbulence cells will cause a fluctuation in the irradiance at the central beam position, a phenomenon known as scintillation. Irradiance fluctuations caused by turbulent atmosphere are also related to the atmospheric structure parameter Cn 2 as noted before. The standard deviation in the irradiance (sI, m) can be estimated for a horizontal beam path [see Eq. (2)] [13],
where k = 2π/λ (m−1) and λ is the laser wavelength (m). From this equation it is concluded that the irradiance fluctuation, and hence the uncertainty of the signal observed, should increase with the range. Less fluctuation is expected with longer laser wavelengths.Atmospheric turbulence may affect to a large extent the ensemble averaging of LIBS spectra. Averaging multiple laser shots on a target in an attempt to increase the LIBS SNR is adversely affected by the combination of beam wander and scintillation phenomena (see below) and the problem gets worse as the range and extent of the turbulence increase. To illustrate the importance of the phenomenon and how LIBS sampling is concerned, a series of 1000 single-shot spectra were acquired from an aluminum target located at 120 m in an open environment. The 1000 shots intended to ablate the same position at the remote target. In the absence of movement, the successive spectra would resemble a depth profile, with the first few spectra showing the aluminum lines plus the coincidental appearance of calcium as a surface contaminant of the aluminum plate. Since calcium is a typical dust component found in most outdoor measurements, calcium surface contamination is detected by the appearance of Ca lines in the first few laser shots delivered on a fresh surface In the presence of wind, calcium contamination is more severe. However, when turbulence occurs the results depart from the expected behavior. Figure 7 shows two spectra arbitrarily chosen from the series. The plot represents raw data. As shown, the top spectrum –corresponding to shot number 481 of the 1000 shot series– exhibits the atomic Al lines at 394.4 nm and 396.1 nm, with no other spectral features present. This means that this shot hit a sample region already ablated –or cleaned– by precedent laser shots. In contrast, the bottom spectrum –corresponding to shot 752, i.e., much deeper into the sample for the case of zero turbulence– shows the Al lines plus the two atomic lines from calcium. This fact indicates that with this shot a fresh surface has been hit rather than a position already ablated by preceding laser shots. This is a typical consequence of beam wander at the target. As shown, the Al intensity also drops when the laser hits a fresh surface. However, this is not a general case as the observed intensities depend on the particular position hit by the beam.
Fig. 7 Two single shot LIBS spectra of Al obtained outdoor at 120 m from the instrument Spectra were arbitrarily chosen from a series of 1000 shots intended to hit a single position on the Al plate.
Figure 8 shows the shot-to-shot variation of the Al and Ca intensities along the 1000 shots series. During the first 100 shots, the general intensity of Al grows as responding to the surface coupling of the incident laser beam. The smaller intensity observed in the first few shots is probably due to the reflectivity of the original surface, and to factors related to the presence of impurities on the sample surface. Beyond that, the average intensity tends to reach a stable level. What is noted in Fig. 8 is that the calcium intensity exhibits sudden increases from time to time, which do not correspond to simultaneous changes in Al intensities. It could be argued that calcium spikes are due to sampling of calcium-rich ambient aerosol particles by the plasma that forms on and above the target surface. However, sampling of dust particles from air is a discrete process and generally is observed as a single shot deviating from the average signal level, not with a number of consecutive shots as depicted in Fig. 8. Our observations associated to the data in Fig. 4 indicate that when the beam deviates from the central position the frequency of Ca spikes increases. These results are compatible with the interpretation proposed linked to beam wander caused by wind gusts. Since the intensity variations of Ca have no correspondence with changes in Al intensity, wind-induced vibration of the system, which could lead to unstable readings for both elements, should be discarded along the full data set. Vibration of the system due to mechanical movement arising from circulating pumps and thermal variations in the laser cavities can be also discarded as our measurements indoor are free from the abrupt effects observed in the open range. Also observed in Fig. 8 is that the Al signal drops in evident coincidence with Ca signal for measurements close to shots 210, 255, 420, 570 and 830. These effects are related to irradiance cutbacks at the target caused by beam spread due to atmospheric turbulence. Other small fluctuations can be ascribed to irradiance changes of smaller amplitude. Changes in mode structure of the lasers in the far field can be also responsible for some of the small fluctuations observed. Data in Fig. 8 were not normalized in order to disclose the original fluctuations.
Fig. 8 Shot-to-shot fluctuation of line intensity from an aluminum target located at 120 m from the instrument (1000 shots). Lines monitored are Al at 394.4 nm and Ca at 393.4 nm.
To have a direct evidence of the atmospheric effects on measurement precision, separate series of spectra were collected outdoor and inside the firing range at 50 m in both instances, while all the instrument parameters (pulse energy, beam focusing, spectrometer settings, etc.) were held constant. The temperature and relative humidity during the two sets of measurements were similar (ca. 18°C, and 37%, respectively). Figure 9 shows the shot-to-shot variation of the Al and Ca intensities of the experiment. As shown, data acquired outdoor show similar trends to those observed in Fig. 8, i.e., the eventual intensity cutbacks persisted. This means that even at 50 m the atmospheric effects due to wind are perceptible. However, the effects caused by the turbulent atmosphere are less appreciable for indoor measurements. Of course, wind was not present inside the corridor. Discarding the first 10 shots in all cases, the precision indoor is 10.8% RSD, while it is 19.0% RSD in the open range. The average intensity observed outdoor (23232 cts.) is slightly smaller than inside the corridor (28349 cts.) due to small losses in pulse energy and plasma emitted light with propagation as a result of particle scattering. Absorption for a range of 50 m by molecular species is negligible at the transmitted laser wavelength (1064 nm), as is also the case for the plasma lines measured in this experiment [14]. The uncertainty observed is 19% RSD at most, which is rather satisfactory for the experimental conditions used.
Fig. 9 Variation of the intensity of Al 394.4 nm and Ca 393.4 nm emission lines with the number of laser pulses at 50 m. a) outdoor and b) indoor.
Table 1 summarizes the statistical parameters associated to single-shot outdoor LIBS measurements on an aluminum target located at different distances from 30 m to 120 m. The signal-to-noise ratio (SNR) was calculated as the ratio of the peak intensity to the root mean square (rms) noise of the peak intensity. As expected, the line intensity decreases with range. Unfortunately the number of measurements is not large enough to establish a reliable fitting equation relating the observed intensity with the range. In a previous paper [9] it was demonstrated to scale at least with the reciprocal fourth power of the range. The data in Table 1 indicate that the signal relative standard deviation increases with the range, an observation consistent with Eq. (2). The SNR also decreases steadily with the propagation distance. However, while the signal drops by 91% when increasing the range from 30 m to 120 m, the SNR only decreases a mere 29%. This means that the signal precision is affected to a lesser extent by the range than the measured peak intensity. The background fluctuation either measured at a single wavelength (408.4 nm) or measured as an average in the featureless wavelength range 408.4-410.0 nm exhibits smaller RSD values with increasing the range. This is certainly to be expected since in the background the rms noise is only proportional to the background count in shot noise limited measurements.
Table 1. Statistical parameters associated to LIBS measurements on an aluminum target located outdoor at variable ranges
4. Conclusions
The results shown here constitute a first quantitative insight into the effect of atmospheric turbulence on TELELIBS measurements using nanosecond lasers. For a system using a focused laser beam induced plasma at the target whose light is recovered by a telescope, the uncertainty observed in outdoor LIBS measurements is determined by beam wander effects of the incident laser beam. Beam wander is primarily the effect of turbulence caused by fluctuations in the index of refraction of the atmosphere and are related to diurnal thermal changes in air properties. Beam wander has important effects on the analytical performance of LIBS for inspection of distant objects. On a heterogeneous target, increasing the SNR by ensemble averaging the signal from multiple laser shots is adversely affected by beam wander and the problem gets more acute as the range and turbulence increase. This problem is particularly perceptible when pointing small targets. If the extent of beam wander exceeds the target size, the analytical capabilities of LIBS are strongly compromised as many hits will fall outside the sample. Over a long time period, beam wander tends to average the sampled area, which is favorable for identification purposes when the analyte of interest is homogeneously distributed across the surface inspected or when the sample is a massive bulk material. Measurement precision tends to be less dependent on the path length than the peak intensity, so a normalization procedure using the background signal as a reference could improve the capabilities of TELELIBS for quantitative analysis. It is clear that TELELIBS emission is detected in the presence of various noise sources, including detector noise, circuit and electronic thermal noise, among other. However, in the presence of adverse atmospheric conditions signal fluctuations caused by irradiance fluctuations and beam wander may become the dominant noise source of TELELIBS measurements.
Inhomogeneities in the temperature and pressure of the atmosphere and aerosol scattering effects caused by fog and other particles have not been considered in this manuscript. Research is in progress to account for these effects. Future tests should include measurements at longer ranges to quantify the effect of haze. Also, the average signal level and signal-level fluctuations should be measured during the experiment so the effect of scintillation on the LIBS measurement can be quantified during adverse weather conditions. This would help to determine whether other losses, such as absorption and multiple scattering, are also significant.
Acknowledgement
This work was supported by Indra Sistemas, S.A. (Madrid) through the research program DeLIBeS. Financial support by the Spain’s Ministerio de Educación y Ciencia (Madrid); project number CTQ2007-60348 and Junta de Andalucia, project number 07-FQM-03308 is also gratefully acknowledged.
References and Links
1. S. Palanco, S. Conesa, and J. J. Laserna, “Analytical control of liquid steel in an induction melting furnace using a remote laser induced plasma spectrometer,” J. Anal. At. Spectrom. 19(4), 462–467 (2004). [CrossRef]
2. U. Panne, R. E. Neuhauser, C. Haisch, H. Fink, and R. Niessner, “Remote analysis of a mineral melt by Laser-Induced Plasma Spectroscopy,” Appl. Spectrosc. 56(3), 375–380 (2002). [CrossRef]
3. A. I. Whitehouse, “Laser-induced breakdown spectroscopy and its application to the remote characterization of hazardous materials,” Spectroscopy Europe 18, 14–21 (2006).
4. R. Grönlund, M. Lundqvist, and S. Svanberg, “Remote imaging laser-induced breakdown spectroscopy and remote cultural heritage ablative cleaning,” Opt. Lett. 30(21), 2882–2884 (2005). [CrossRef] [PubMed]
5. B. Sallé, P. Mauchien, and S. Maurice, “Laser-Induced Breakdown Spectroscopy in open-path configuration for the analysis of distant objects,” Spectrochim. Acta, B At. Spectrosc. 62(8), 739–768 (2007). [CrossRef]
6. J. J. Laserna, in First International Conference on Laser Induced Breakdown Spectroscopy and Applications (LIBS 2000), (personal communication, 2000).
7. S. Palanco, J. M. Baena, and J. J. Laserna, “Open-path laser-induced plasma spectrometry for remote analytical measurements on solid surfaces,” Spectrochim. Acta, B At. Spectrosc. 57(3), 591–599 (2002). [CrossRef]
8. J. M. Vadillo, P. L. García, S. Palanco, D. Romero, J. M. Baena, and J. J. Laserna, “Remote, real-time, on-line monitoring of high-temperature samples by noninvasive open-path laser plasma spectrometry,” Anal. Bioanal. Chem. 375(8), 1144–1147 (2003). [PubMed]
9. R. Grönlund, M. Lundqvist, and S. Svanberg, “Remote imaging laser-induced breakdown spectroscopy and laser-induced fluorescence spectroscopy using nanosecond pulses from a mobile lidar system,” Appl. Spectrosc. 60(8), 853–859 (2006). [CrossRef] [PubMed]
10. S. Palanco, C. Lopez Moreno, and J. J. Laserna, “Design, construction and assessment of a field-deployable laser-induced breakdown spectrometer for remote elemental sensing,” Spectrochim. Acta, B At. Spectrosc. 61(1), 88–95 (2006). [CrossRef]
11. L. C. Andrews, and R. L. Phillips, Laser Beam Propagation through Random Media 2nd Ed (SPIE Press, Bellingham, 2005).
12. A. Zilberman, N. S. Kopeida, and Y. Sorani, Laser beam widening as a function of elevation in the atmosphere for horizontal propagation, Proc. SPIE Laser Weapons Technology II, vol. 4376, 2001.
13. E. Friedman, and J. L. Miller, Photonics Rules of Thumb 2nd. Ed. (McGraw-Hill New York, 2004), pp. 176.
14. A. Ferrero and J. J. Laserna, “A theoretical study of atmospheric propagation of laser and return light for stand-off laser induced breakdown spectroscopy purposes,” Spectrochimica Acta Part B 63(2), 305–311 (2008). [CrossRef]
